import pandas as pd
import numpy as np

df = pd.read_csv('/content/banana_quality.csv')

df

	Size	Weight	Sweetness	Softness	HarvestTime	Ripeness	Acidity	Quality
0	-1.924968	0.468078	3.077832	-1.472177	0.294799	2.435570	0.271290	Good
1	-2.409751	0.486870	0.346921	-2.495099	-0.892213	2.067549	0.307325	Good
2	-0.357607	1.483176	1.568452	-2.645145	-0.647267	3.090643	1.427322	Good
3	-0.868524	1.566201	1.889605	-1.273761	-1.006278	1.873001	0.477862	Good
4	0.651825	1.319199	-0.022459	-1.209709	-1.430692	1.078345	2.812442	Good
...	...	...	...	...	...	...	...	...
7995	-6.414403	0.723565	1.134953	2.952763	0.297928	-0.156946	2.398091	Bad
7996	0.851143	-2.217875	-2.812175	0.489249	-1.323410	-2.316883	2.113136	Bad
7997	1.422722	-1.907665	-2.532364	0.964976	-0.562375	-1.834765	0.697361	Bad
7998	-2.131904	-2.742600	-1.008029	2.126946	-0.802632	-3.580266	0.423569	Bad
7999	-2.660879	-2.044666	0.159026	1.499706	-1.581856	-1.605859	1.435644	Bad

8000 rows × 8 columns

Distributions

2-d distributions

Time series

Values

numerical = []
categorical = []
for column in df.columns:
  if df[column].dtype!='O':
    numerical.append(column)
  else:
    categorical.append(column)

print(numerical)
print(categorical)

['Size', 'Weight', 'Sweetness', 'Softness', 'HarvestTime', 'Ripeness', 'Acidity']
['Quality']

df.isnull().sum()

Size           0
Weight         0
Sweetness      0
Softness       0
HarvestTime    0
Ripeness       0
Acidity        0
Quality        0
dtype: int64

unique_values = {}
for feature in categorical:
    unique_values[feature] = df[feature].unique()

df_unique_values = pd.DataFrame.from_dict([unique_values]).transpose().reset_index()

df_unique_values.columns = ['Categorical Feature','Unique_values']
df_unique_values['Number of Unique Values'] = df_unique_values['Unique_values'].apply(lambda x: len(x))

df_unique_values

	Categorical Feature	Unique_values	Number of Unique Values
0	Quality	[Good, Bad]	2

df['Quality'] = df['Quality'].replace({'Good': 0, 'Bad': 1})

df

	Size	Weight	Sweetness	Softness	HarvestTime	Ripeness	Acidity	Quality
0	-1.924968	0.468078	3.077832	-1.472177	0.294799	2.435570	0.271290	0
1	-2.409751	0.486870	0.346921	-2.495099	-0.892213	2.067549	0.307325	0
2	-0.357607	1.483176	1.568452	-2.645145	-0.647267	3.090643	1.427322	0
3	-0.868524	1.566201	1.889605	-1.273761	-1.006278	1.873001	0.477862	0
4	0.651825	1.319199	-0.022459	-1.209709	-1.430692	1.078345	2.812442	0
...	...	...	...	...	...	...	...	...
7995	-6.414403	0.723565	1.134953	2.952763	0.297928	-0.156946	2.398091	1
7996	0.851143	-2.217875	-2.812175	0.489249	-1.323410	-2.316883	2.113136	1
7997	1.422722	-1.907665	-2.532364	0.964976	-0.562375	-1.834765	0.697361	1
7998	-2.131904	-2.742600	-1.008029	2.126946	-0.802632	-3.580266	0.423569	1
7999	-2.660879	-2.044666	0.159026	1.499706	-1.581856	-1.605859	1.435644	1

8000 rows × 8 columns

df[numerical[:]].hist(figsize=(10,10),bins=100)

array([[<Axes: title={'center': 'Size'}>,
        <Axes: title={'center': 'Weight'}>,
        <Axes: title={'center': 'Sweetness'}>],
       [<Axes: title={'center': 'Softness'}>,
        <Axes: title={'center': 'HarvestTime'}>,
        <Axes: title={'center': 'Ripeness'}>],
       [<Axes: title={'center': 'Acidity'}>, <Axes: >, <Axes: >]],
      dtype=object)

import seaborn as sns

# Combine the features and target variable into one DataFrame
df_combined = X_train.copy()
df_combined['Quality'] = y_train

# Plot scatter plots for every pair of features against each other
sns.pairplot(df_combined, hue='Quality', diag_kind='kde')
plt.show()

import seaborn as sns
import matplotlib.pyplot as plt

# Plot distribution plots for each feature
for feature in X_train.columns:
    sns.displot(X_train[feature], kde=True)
    plt.title(f'Distribution of {feature}')
    plt.xlabel(feature)
    plt.ylabel('Density')
    plt.show()

df_scaled = df.copy()
for column in numerical:
    df_scaled[column] = (df[column] - df[column].min()) / (df[column].max() - df[column].min())

df_scaled

	Size	Weight	Sweetness	Softness	HarvestTime	Ripeness	Acidity	Quality
0	0.405073	0.599374	0.911017	0.384425	0.604782	0.640332	0.518495	0
1	0.363736	0.600941	0.628708	0.302919	0.491862	0.607889	0.521215	0
2	0.538722	0.683992	0.754984	0.290963	0.515164	0.698081	0.605769	0
3	0.495156	0.690913	0.788183	0.400234	0.481011	0.590738	0.534090	0
4	0.624796	0.670323	0.590524	0.405338	0.440636	0.520684	0.710337	0
...	...	...	...	...	...	...	...	...
7995	0.022259	0.620671	0.710171	0.737001	0.605080	0.411786	0.679056	1
7996	0.641792	0.375475	0.302137	0.540710	0.450842	0.221374	0.657544	1
7997	0.690530	0.401334	0.331062	0.578615	0.523239	0.263876	0.550661	1
7998	0.387428	0.331734	0.488640	0.671200	0.500384	0.109999	0.529991	1
7999	0.342322	0.389914	0.609285	0.621222	0.426256	0.284055	0.606397	1

8000 rows × 8 columns

# import the necessary libraries
import seaborn as sns
import matplotlib.pyplot as plt

# to ignore the warnings
from warnings import filterwarnings

for feature in numerical:
  if feature!='index':
      fig = plt.figure()
      plt.xlabel("feature")
      sns.distplot(df[feature])

<ipython-input-10-304298f86003>:12: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(df[feature])
<ipython-input-10-304298f86003>:12: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(df[feature])
<ipython-input-10-304298f86003>:12: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(df[feature])
<ipython-input-10-304298f86003>:12: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(df[feature])
<ipython-input-10-304298f86003>:12: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(df[feature])
<ipython-input-10-304298f86003>:12: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(df[feature])
<ipython-input-10-304298f86003>:12: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(df[feature])

# import the necessary libraries
import seaborn as sns
import matplotlib.pyplot as plt

# to ignore the warnings
from warnings import filterwarnings

for feature in numerical:
  if feature!='index':
      fig = plt.figure()
      sns.boxplot(df[feature])

import matplotlib.pyplot as plt
for column in numerical:
    Q1 = df[column].quantile(.25)
    Q3 = df[column].quantile(.75)
    IQR = Q3 - Q1
    UB = Q3 + 1.5 * IQR
    LB = Q1 - 1.5 * IQR
    df.loc[df[column] >= UB, column] = UB
    df.loc[df[column] <= LB, column] = LB
    plt.figure()
    df[[column]].boxplot()

Model Training and Testing

import numpy

from sklearn.model_selection import train_test_split

X = df.drop(['Quality'] ,axis=1)
y = df['Quality']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

import numpy as np
from sklearn.model_selection import GridSearchCV
from sklearn.linear_model import LogisticRegression
from sklearn.neighbors import KNeighborsClassifier
from sklearn.naive_bayes import GaussianNB
from sklearn.cluster import KMeans
from sklearn.pipeline import Pipeline
from sklearn.tree import DecisionTreeClassifier

# Define models
models = [
    {
        'name': 'Logistic Regression',
        'model': LogisticRegression(),
        'params': {}
    },
    {
        'name': 'KNN',
        'model': KNeighborsClassifier(),
        'params': {
            'n_neighbors': [3, 5, 7, 9]
        }
    },
    {
        'name': 'Gaussian Naive Bayes',
        'model': GaussianNB(),
        'params': {}
    },
    {
        'name': 'Decision Tree',
        'model': DecisionTreeClassifier(),
        'params': {
            'max_depth': [None, 5, 10, 15, 20]
        }
    }
]

# Perform GridSearchCV for each model
for model_info in models:
    print(f"Performing GridSearchCV for {model_info['name']}...")
    grid_search = GridSearchCV(model_info['model'], model_info['params'], cv=5)
    grid_search.fit(X_train, y_train)
    print("Best parameters found:", grid_search.best_params_)
    print("Best score found:", grid_search.best_score_)
    print("\n")

Performing GridSearchCV for Logistic Regression...
Best parameters found: {}
Best score found: 0.8768750000000001


Performing GridSearchCV for KNN...
Best parameters found: {'n_neighbors': 7}
Best score found: 0.9806250000000001


Performing GridSearchCV for Gaussian Naive Bayes...
Best parameters found: {}
Best score found: 0.885


Performing GridSearchCV for Decision Tree...
Best parameters found: {'max_depth': 10}
Best score found: 0.93484375

from sklearn.metrics import accuracy_score

# Create KNN model with n_neighbors=5
knn = KNeighborsClassifier(n_neighbors=5)

# Fit the model
knn.fit(X_train, y_train)

# Predict on the test data
y_pred = knn.predict(X_test)

# Calculate accuracy score
accuracy = accuracy_score(y_test, y_pred)
print("Accuracy Score:", accuracy)

Accuracy Score: 0.980625

from sklearn.metrics import confusion_matrix, accuracy_score, roc_auc_score, roc_curve

# Assuming y_Test and y_pred are your true labels and predicted labels respectively

# Confusion Matrix
cm = confusion_matrix(y_test, y_pred)
print("Confusion Matrix:")
print(cm)

Confusion Matrix:
[[808  11]
 [ 11 770]]

sns.heatmap(cm,
            annot=True,
            fmt='g',
            xticklabels=['0','1'],
            yticklabels=['0','1'])
plt.ylabel('Prediction',fontsize=13)
plt.xlabel('Actual',fontsize=13)
plt.title('Confusion Matrix',fontsize=17)
plt.show()

def plot_roc(y_Test,y_pred):
   fpr,tpr,thresholds = roc_curve(y_Test,y_pred)
   plt.plot(fpr,tpr)
   plt.xlabel("False positive rate")
   plt.ylabel("True positive rate")
   plt.show()

plot_roc(y_test,y_pred)

import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Dropout
import matplotlib.pyplot as plt
from sklearn.metrics import confusion_matrix
import seaborn as sns
# Define the model
model = Sequential([
    Dense(512, activation='relu', input_shape=(X_train.shape[1],)),  # Hidden layer with 64 units and ReLU activation
    Dense(256,activation='relu'),
    Dropout(0.2),  # Dropout layer with a dropout rate of 0.2
    Dense(1, activation='sigmoid')  # Output layer with 1 unit and sigmoid activation for binary classification
])

# Compile the model
model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])

# Fit the model
history = model.fit(X_train, y_train, epochs=50, batch_size=32, validation_data=(X_test, y_test))

# Evaluate the model
loss, accuracy = model.evaluate(X_test, y_test)
print("Loss:", loss)
print("Accuracy:", accuracy)


# Plot accuracy
plt.plot(history.history['accuracy'], label='Training Accuracy')
plt.plot(history.history['val_accuracy'], label='Testing Accuracy')
plt.title('Model Accuracy')
plt.xlabel('Epoch')
plt.ylabel('Accuracy')
plt.legend()
plt.show()

# Plot loss
plt.plot(history.history['loss'], label='Training Loss')
plt.plot(history.history['val_loss'], label='Testing Loss')
plt.title('Model Loss')
plt.xlabel('Epoch')
plt.ylabel('Loss')
plt.legend()
plt.show()

# Predict on test data
y_pred = model.predict(X_test)
y_pred_classes = (y_pred > 0.5).astype(int)

# Create confusion matrix
cm = confusion_matrix(y_test, y_pred_classes)

# Plot confusion matrix
plt.figure(figsize=(8, 6))
sns.heatmap(cm, annot=True, fmt='d', cmap='Blues')
plt.xlabel('Predicted labels')
plt.ylabel('True labels')
plt.title('Confusion Matrix')
plt.show()

Epoch 1/50
200/200 [==============================] - 2s 7ms/step - loss: 0.1440 - accuracy: 0.9498 - val_loss: 0.0869 - val_accuracy: 0.9725
Epoch 2/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0989 - accuracy: 0.9711 - val_loss: 0.0910 - val_accuracy: 0.9675
Epoch 3/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0888 - accuracy: 0.9727 - val_loss: 0.0823 - val_accuracy: 0.9806
Epoch 4/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0840 - accuracy: 0.9772 - val_loss: 0.0756 - val_accuracy: 0.9769
Epoch 5/50
200/200 [==============================] - 1s 7ms/step - loss: 0.0827 - accuracy: 0.9758 - val_loss: 0.0728 - val_accuracy: 0.9781
Epoch 6/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0801 - accuracy: 0.9777 - val_loss: 0.0773 - val_accuracy: 0.9781
Epoch 7/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0739 - accuracy: 0.9784 - val_loss: 0.0784 - val_accuracy: 0.9794
Epoch 8/50
200/200 [==============================] - 2s 8ms/step - loss: 0.0753 - accuracy: 0.9798 - val_loss: 0.0713 - val_accuracy: 0.9812
Epoch 9/50
200/200 [==============================] - 2s 8ms/step - loss: 0.0720 - accuracy: 0.9797 - val_loss: 0.0832 - val_accuracy: 0.9756
Epoch 10/50
200/200 [==============================] - 1s 6ms/step - loss: 0.0688 - accuracy: 0.9797 - val_loss: 0.0680 - val_accuracy: 0.9831
Epoch 11/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0682 - accuracy: 0.9812 - val_loss: 0.0677 - val_accuracy: 0.9806
Epoch 12/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0655 - accuracy: 0.9820 - val_loss: 0.0669 - val_accuracy: 0.9825
Epoch 13/50
200/200 [==============================] - 1s 6ms/step - loss: 0.0676 - accuracy: 0.9812 - val_loss: 0.0670 - val_accuracy: 0.9844
Epoch 14/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0628 - accuracy: 0.9803 - val_loss: 0.0694 - val_accuracy: 0.9819
Epoch 15/50
200/200 [==============================] - 1s 6ms/step - loss: 0.0639 - accuracy: 0.9822 - val_loss: 0.0693 - val_accuracy: 0.9819
Epoch 16/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0596 - accuracy: 0.9834 - val_loss: 0.0718 - val_accuracy: 0.9794
Epoch 17/50
200/200 [==============================] - 1s 6ms/step - loss: 0.0587 - accuracy: 0.9816 - val_loss: 0.0670 - val_accuracy: 0.9850
Epoch 18/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0565 - accuracy: 0.9830 - val_loss: 0.0646 - val_accuracy: 0.9837
Epoch 19/50
200/200 [==============================] - 2s 8ms/step - loss: 0.0537 - accuracy: 0.9844 - val_loss: 0.0789 - val_accuracy: 0.9812
Epoch 20/50
200/200 [==============================] - 1s 7ms/step - loss: 0.0564 - accuracy: 0.9830 - val_loss: 0.0642 - val_accuracy: 0.9844
Epoch 21/50
200/200 [==============================] - 1s 6ms/step - loss: 0.0540 - accuracy: 0.9830 - val_loss: 0.0766 - val_accuracy: 0.9831
Epoch 22/50
200/200 [==============================] - 1s 6ms/step - loss: 0.0562 - accuracy: 0.9841 - val_loss: 0.0649 - val_accuracy: 0.9837
Epoch 23/50
200/200 [==============================] - 1s 6ms/step - loss: 0.0509 - accuracy: 0.9858 - val_loss: 0.0856 - val_accuracy: 0.9787
Epoch 24/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0512 - accuracy: 0.9866 - val_loss: 0.0693 - val_accuracy: 0.9831
Epoch 25/50
200/200 [==============================] - 1s 6ms/step - loss: 0.0506 - accuracy: 0.9855 - val_loss: 0.0726 - val_accuracy: 0.9825
Epoch 26/50
200/200 [==============================] - 1s 6ms/step - loss: 0.0467 - accuracy: 0.9862 - val_loss: 0.0773 - val_accuracy: 0.9800
Epoch 27/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0477 - accuracy: 0.9859 - val_loss: 0.0732 - val_accuracy: 0.9812
Epoch 28/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0459 - accuracy: 0.9858 - val_loss: 0.0784 - val_accuracy: 0.9831
Epoch 29/50
200/200 [==============================] - 1s 7ms/step - loss: 0.0440 - accuracy: 0.9861 - val_loss: 0.0836 - val_accuracy: 0.9837
Epoch 30/50
200/200 [==============================] - 2s 8ms/step - loss: 0.0470 - accuracy: 0.9861 - val_loss: 0.0944 - val_accuracy: 0.9800
Epoch 31/50
200/200 [==============================] - 1s 4ms/step - loss: 0.0438 - accuracy: 0.9878 - val_loss: 0.0749 - val_accuracy: 0.9837
Epoch 32/50
200/200 [==============================] - 1s 6ms/step - loss: 0.0441 - accuracy: 0.9869 - val_loss: 0.0859 - val_accuracy: 0.9812
Epoch 33/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0436 - accuracy: 0.9867 - val_loss: 0.0807 - val_accuracy: 0.9844
Epoch 34/50
200/200 [==============================] - 1s 6ms/step - loss: 0.0397 - accuracy: 0.9872 - val_loss: 0.0903 - val_accuracy: 0.9837
Epoch 35/50
200/200 [==============================] - 1s 4ms/step - loss: 0.0422 - accuracy: 0.9867 - val_loss: 0.0825 - val_accuracy: 0.9856
Epoch 36/50
200/200 [==============================] - 1s 6ms/step - loss: 0.0386 - accuracy: 0.9884 - val_loss: 0.1002 - val_accuracy: 0.9844
Epoch 37/50
200/200 [==============================] - 1s 6ms/step - loss: 0.0375 - accuracy: 0.9892 - val_loss: 0.0848 - val_accuracy: 0.9825
Epoch 38/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0369 - accuracy: 0.9878 - val_loss: 0.0875 - val_accuracy: 0.9831
Epoch 39/50
200/200 [==============================] - 2s 8ms/step - loss: 0.0387 - accuracy: 0.9873 - val_loss: 0.0934 - val_accuracy: 0.9837
Epoch 40/50
200/200 [==============================] - 2s 9ms/step - loss: 0.0372 - accuracy: 0.9886 - val_loss: 0.0922 - val_accuracy: 0.9837
Epoch 41/50
200/200 [==============================] - 1s 6ms/step - loss: 0.0367 - accuracy: 0.9884 - val_loss: 0.0908 - val_accuracy: 0.9831
Epoch 42/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0339 - accuracy: 0.9894 - val_loss: 0.1027 - val_accuracy: 0.9837
Epoch 43/50
200/200 [==============================] - 1s 6ms/step - loss: 0.0332 - accuracy: 0.9903 - val_loss: 0.1135 - val_accuracy: 0.9831
Epoch 44/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0322 - accuracy: 0.9895 - val_loss: 0.0974 - val_accuracy: 0.9844
Epoch 45/50
200/200 [==============================] - 1s 6ms/step - loss: 0.0309 - accuracy: 0.9897 - val_loss: 0.0987 - val_accuracy: 0.9837
Epoch 46/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0335 - accuracy: 0.9887 - val_loss: 0.1045 - val_accuracy: 0.9831
Epoch 47/50
200/200 [==============================] - 1s 6ms/step - loss: 0.0310 - accuracy: 0.9898 - val_loss: 0.1131 - val_accuracy: 0.9812
Epoch 48/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0312 - accuracy: 0.9898 - val_loss: 0.0995 - val_accuracy: 0.9862
Epoch 49/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0334 - accuracy: 0.9894 - val_loss: 0.1090 - val_accuracy: 0.9837
Epoch 50/50
200/200 [==============================] - 1s 5ms/step - loss: 0.0326 - accuracy: 0.9894 - val_loss: 0.1106 - val_accuracy: 0.9831
50/50 [==============================] - 0s 3ms/step - loss: 0.1106 - accuracy: 0.9831
Loss: 0.11059460788965225
Accuracy: 0.9831249713897705

50/50 [==============================] - 0s 2ms/step

##Save KNN Model
import pickle
with open('banana_quilty.pkl','wb') as f:
    pickle.dump(model,f)

from joblib import dump
dump(model, 'banana_quilty.joblib')

['banana_quilty.joblib']

model.save('banana_model.h5')

/usr/local/lib/python3.10/dist-packages/keras/src/engine/training.py:3103: UserWarning: You are saving your model as an HDF5 file via `model.save()`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')`.
  saving_api.save_model(

from keras.models import load_model
model = load_model('banana_model.h5')

import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.neural_network import MLPClassifier
import joblib

# Load the dataset
data = df

# Split features and target variable
# X = data.drop('Quality', axis=1)  # Replace 'Quality' with the actual name of your target column
# y = data['Quality']

# Split the dataset into training and testing sets
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# Standardize features by removing the mean and scaling to unit variance
scaler = StandardScaler()
X_train_scaled = scaler.fit_transform(X_train)
X_test_scaled = scaler.transform(X_test)

# Initialize and train the MLPClassifier
model = MLPClassifier(hidden_layer_sizes=(100, 50), max_iter=500, random_state=42)
model.fit(X_train_scaled, y_train)

# Evaluate the model
accuracy = model.score(X_test_scaled, y_test)
print("Accuracy:", accuracy)

# Save the trained model using joblib
joblib.dump(model, 'banana_quilty.joblib')

import matplotlib.pyplot as plt
from sklearn.metrics import confusion_matrix, classification_report

# Predictions on the test set
y_pred = model.predict(X_test_scaled)

# Confusion Matrix
cm = confusion_matrix(y_test, y_pred)
print("Confusion Matrix:")
print(cm)

# Plot Confusion Matrix
plt.imshow(cm, interpolation='nearest', cmap=plt.cm.Blues)
plt.title('Confusion Matrix')
plt.colorbar()
plt.xlabel('Predicted')
plt.ylabel('Actual')
plt.show()

# Classification Report
print("Classification Report:")
print(classification_report(y_test, y_pred))

Accuracy: 0.98625
Confusion Matrix:
[[808  11]
 [ 11 770]]

Classification Report:
              precision    recall  f1-score   support

           0       0.99      0.99      0.99       819
           1       0.99      0.99      0.99       781

    accuracy                           0.99      1600
   macro avg       0.99      0.99      0.99      1600
weighted avg       0.99      0.99      0.99      1600